Growth and specific P-uptake rates of bacterial and phytoplanktonic communities in the Southeast Pacific (BIOSOPE cruise)

(1)

HAL Id: hal-00330248

https://hal.archives-ouvertes.fr/hal-00330248

Submitted on 27 Jun 2007

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Growth and specific P-uptake rates of bacterial and

phytoplanktonic communities in the Southeast Pacific

(BIOSOPE cruise)

S. Duhamel, T. Moutin, France van Wambeke, B. van Mooy, P. Rimmelin,

Patrick Raimbault, Hervé Claustre

To cite this version:

S. Duhamel, T. Moutin, France van Wambeke, B. van Mooy, P. Rimmelin, et al.. Growth and specific P-uptake rates of bacterial and phytoplanktonic communities in the Southeast Pacific (BIOSOPE cruise). 2007, pp.2027-2068. �hal-00330248�

(2)

BGD

4, 2027–2068, 2007

Specific P-uptake rates of bacteria and

phytoplankton S. Duhamel et al. Title Page Abstract Introduction Conclusions References Tables Figures ◭ ◮ ◭ ◮ Back Close

Full Screen / Esc

Printer-friendly Version Interactive Discussion

EGU Biogeosciences Discuss., 4, 2027–2068, 2007

www.biogeosciences-discuss.net/4/2027/2007/ © Author(s) 2007. This work is licensed

under a Creative Commons License.

Biogeosciences Discussions

Biogeosciences Discussions is the access reviewed discussion forum of Biogeosciences

Growth and specific P-uptake rates of

bacterial and phytoplanktonic

communities in the Southeast Pacific

(BIOSOPE cruise)

S. Duhamel1,2, T. Moutin1, F. Van Wambeke2, B. Van Mooy3, P. Rimmelin1, P. Raimbault1, and H. Claustre4

1

Aix-Marseille Universit é, Laboratoire d’Oc éanographie et de Biog éochimie, LOB-UMR 6535 CNRS, OSU/Centre d’Oc éanologie de Marseille, 13288 Marseille, Cedex 09, France

2

Aix-Marseille Universit é, Laboratoire de Microbiologie, G éochimie et Ecologie Marines, LMGEM-UMR 6117 CNRS, OSU/Centre d’Oc éanologie de Marseille, 13288 Marseille, Cedex 09, France

3

Woods Hole Oceanographic Institution, Mailstop 4 Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA

4

Observatoire Oc ´eanologique de Villefranche, Laboratoire d’Oc ´eanographie de Villefranche, UMR7093, B.P. 08, 06238 Villefranche-sur-mer, France

Received: 15 June 2007 – Accepted: 15 June 2007 – Published: 27 June 2007 Correspondence to: S. Duhamel (solange.duhamel@univmed.fr)

(3)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU

Abstract

Predicting heterotrophic bacteria and phytoplankton growth rates (µ) is of great

sci-entific interest. Many methods have been developed in order to assess bacterial or phytoplankton µ. One widely used method is to estimate µ from data obtained on

biomass or cell abundance and rates of biomass or cell production. According to

5

Kirchman (2002), the most appropriate approach for estimatingµ is simply to divide

the production rate by the biomass or cell abundance estimate. Most of the methods using this approach are expressed using carbon (C) data. Nevertheless it is also pos-sible to estimateµ using phosphate (P) data. We showed that particulate phosphate

(PartP) can be used to estimate biomass and that the phosphate uptake rate to PartP

10

ratio can be employed to assessµ. Contrary to other methods using C, this estimator

does not need conversion factors and provides an evaluation ofµ for both autotrophic

and heterotrophic organisms. We report values of P-based µ in three size fractions

(0.2–0.6; 0.6–2 and>2 µm) along a Southeast Pacific transect, over a wide range of

P-replete trophic status. P-basedµ values were higher in the 0.6–2 µm fraction than

15

in the >2 µm fraction, suggesting that picoplankton-sized cells grew faster than the

larger cells, whatever the trophic regime encountered. Picoplankton-sized cells grew significantly faster in the deep chlorophyll maximum layer than in the upper part of the photic zone in the oligotrophic gyre area, suggesting that picoplankton might outcom-pete>2 µm cells in this particular high-nutrient, low-light environment. P-based µ

at-20

tributed to free-living bacteria (0.2–0.6µm) and picoplankton (0.6–2 µm) size-fractions

were relatively low (0.11±0.07 d−1 and 0.14±0.04 d−1, respectively) in the Southeast Pacific gyre, suggesting that the microbial community turns over very slowly.

1 Introduction

A fundamental aim in ecology and hence, biological oceanography and limnology, is

25

(4)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU (Banse, 2002). An assessment of the ecological role of both autotrophic and

het-erotrophic marine micro-organisms depends, to a significant extent, on estimates of their µ (Azam et al., 1983). Phytoplankton µ estimates vary widely from values of

around 0.1–0.3 d−1 (Letelier et al., 1996; Maranon et al., 2000, 2005) to 1–2 d−1(Laws et al., 1987; Quevedo and Anadon, 2001). Bacterialµ estimates also vary widely, from

5

very low values 0.004–0.25 d−1 (Sherr et al., 2001; Van Wambeke, 2007b1) to higher values of around 2–10 d−1 (Ducklow, 1983; Jones et al., 1996). Studies comparing bacterial and phytoplankton _{µ are scarce and show significant differences between} these organisms (Jones et al., 1996). The determination of heterotrophic bacterial and phytoplanktonµ is critical in order to predict the responses of the planktonic ecosystem

10

to potential changes in nutrient supply to the upper ocean.

Numerous methods have been developed to measure µ (Brock, 1971). In order to

estimate phytoplankton and heterotrophic bacterialµ, authors have used various direct

or indirect methodologies of varying accuracy. The two most common direct methods, applicable to both heterotrophic bacteria and phytoplankton, are the frequency of

di-15

viding cells (Hagstrom et al., 1979) and the dilution technique (Landry and Hassett, 1982; Quevedo and Anadon, 2001). Direct methods are difficult to set up on board so microbial growth rates are commonly calculated from production and standing stock data. According to Kirchman (2002), the most appropriate approach to estimateµ of

microbial assemblages is the simplest, that is, to divide the production rate by the

es-20

timate of biomass (B) or cell abundance. This ratio is called the “specific uptake rate” (Vsp) (Lipschultz, 1995; Dickson and Wheeler, 1995) or specific growth rate (Laws et al., 1984; Mara ˜non et al., 2003; Holl and Montoya, 2005) and it is an expression of the

µ, as it is modified by resource limitation, temperature and predation (Brock, 1971).

The most common indirect methods to measure phytoplanktonµ are14C-pigment

la-25

belling (Redalje and Laws, 1981; Welschmeyer et al., 1991; Jones et al., 1996; Cailliau

1

Van Wambeke, F., Obernosterer, I., Moutin, T., Duhamel, S., Ulloa, O., and Claustre, H.: Heterotrophic prokaryotic production in the South East Pacific: longitudinal trends and coupling with primary production, Biogeosciences Discuss., in preparation, 2007b.

(5)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU et al., 1996), cell cycle analysis (Vaulot, 1992; Liu et al., 1999) and the use of

equa-tions linking autotrophic production (AP) and autotrophic cell abundance or biomass (AB) (Smith et al., 2000; Maranon, 2005). In such equations, AP is deduced from the NaH14CO3 incorporation uptake rate measurements (Steemann-Nielsen, 1951). The most common indirect method for studying heterotrophic bacterialµ, is the use of

equa-5

tions that link heterotrophic bacterial production (HBP) and heterotrophic bacterial cell abundance or biomass (HBB). In such equations, HBP is generally deduced from the incorporation of3H-thymidine (Fuhrman and Azam, 1980, 1982) and3H-leucine (Kirch-man et al., 1985) into DNA and proteins respectively. More recently, measurements of the incorporation rates of33PO4 into phospholipids (particularly phosphatidylglycerol:

10

PG and phosphatidylethanolamine: PE) specific to bacterioplankton have been used (Van Mooy et al., 2006).

In numerous studies, authors have estimated phytoplanktonµ by dividing AP,

mea-sured using the 14C method, by various AB estimators such as Chlorophyll a (Chla), particulate organic carbon (POC) and carbon (C) content estimated from microscopy

15

or flow cytometry measurements (Eppley, 1972; Vadstein et al., 1988; Malone et al., 1993; Maranon et al., 2000, 2005; Moreira-Turcq, 2001). The use of Chla and POC as AB proxies is debatable (Le Floc’h et al., 2002; Sobczak et al., 2002; Huot et al., this issue) and C content estimates are dependant on conversion factors. These con-version factors can vary greatly between studies. In the same way, the evaluation

20

of bacterial µ based on the HBP to HBB ratio requires several conversion factors (to

convert the incorporation of 3H-leucine or 3H-thymidine to C equivalents and to con-vert cell number to biomass equivalents) varying with different studies (Riemann et al., 1990). Furthermore, method comparisons can show significant differences between µ estimates (Laws et al., 1984).

25

If_{µ = production/biomass, then there is a direct relationship between incorporation} rate per cell andµ (Kirchman, 2002). Although biomass and production estimators are

usually expressed in terms of C, it is also possible to express them in terms of nitro-gen (N) or phosphate (P) as C, N and P are major cellular constituents linked via the

(6)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU “Redfield ratio” (Redfield, 1963; Berman, 1980; Laws et al., 1984). Contrary to C and

N, P is more quickly liberated from dead material (Menzel and Ryther, 1964; Knauer et al., 1979; Minster and Boulahdid, 1987; Clark et al., 1999). As a consequence, in the open ocean, the proportion of detrital P in PartP is low (Faul et al., 2005). Phosphate uptake rates are commonly measured using the32P or33P method and quantifies the

5

amount of P that is taken up by both heterotrophic and autotrophic cells. Measuring the dissolved inorganic P (DIP) uptake rates provides an estimate for planktonic produc-tion, assuming DIP is the sole source of P and there is no, or negligible luxury uptake (Thingstad et al., 1996). Thus, particulate P (PartP) and P uptake rate can be used as estimators of planktonic biomass and production, respectively.

10

We estimated µ from production to biomass ratios expressed in terms of P and

discussed the bias associated with using C and P-based µ estimations. Combining

P uptake rates and PartP measurements with size fractionations, we determined the DIP specific uptake rate (V_DIPsp) in three size fractions corresponding to heterotrophic bacteria, picophytoplankton and nano-microphytoplankton (0.2–0.6; 0.6–2 and>2 µm,

15

respectively), following an east-west transect along the Southeast Pacific ocean. Yet, this area present an interesting diversity of trophic conditions among which the South-east Pacific gyre which is the largest, lSouth-east described and most oligotrophic anticy-clonic gyre of the ocean (Claustre and Maritorena, 2003; Claustre et al., 20072). The measurement ofV_DIPsp in the different compartments enabled us to compare bacterial to

20

phytoplanktonµ using the same method and to study the variability of dynamics among

2 major groups of phytoplankton.

2

Claustre, H., Sciandra, A., and Vaulot, D.: Introduction to the special section: bio-optical and biogeochemical conditions in the South East Pacific in late 2004 – the BIOSOPE cruise, Biogeosciences Discuss., in preparation, 2007.

(7)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU

2 Materials and methods

2.1 Station locations, sample collection and hydrological characteristics

This work was conducted during the BIOSOPE (BIogeochemistry and Optics SOuth Pacific Experiment) cruise in the Southeast Pacific Ocean (between 146.36◦W and 72.49◦W; Fig. 1). The cruise was carried out aboard “l’Atalante” from October to

5

December 2004. High vertical resolution environmental data was produced using a conductivity temperature-depth-oxygen profiler (CTDO, Seabird 911 Plus), from 0 and 500 m, measuring external temperature, conductivity, salinity, oxygen, fluorescence and depth (see Claustre et al., 20072; Ras et al., 20073, for hydrodynamical entities, hydrographic conditions and pigment distribution). Seawater samples were collected

10

at 6 predetermined depths corresponding to 6 levels of the Photosynthetically Active Radiations (PAR – 50, 25, 15, 7, 3 and 1% of surface irradiance). Samples were col-lected in 12 L Niskin bottles attached to a rosette CTD system, at 09:00 a.m. (local time). Subsamples were collected directly, without pre-filtration, into clean, sample-rinsed polycarbonate bottles.

15

2.2 Analytical methods

Particulate phosphate (PartP) was measured using the Strickland and Parsons pro-cedure (1972) for standard DIP, following high-temperature persulfate wet-oxidation at 120◦_{C and 1 bar (Pujo-Pay and Raimbault, 1994). Sequential filtration was carried out} on 1 to 1.2 L samples through different porosity polycarbonate filters (0.2, 0.6, and

20

2µm; 47 mm) using Sartorius systems and very low vacuum (drop by drop). The 0.2

and 0.6µm filters in the lower Sartorius system were separated by a nylon separator

(NY8H04700, Millipore) previously treated by persulfate wet-oxidation to lower blank values. Immediately after filtration, the filters (and the separator for the 0.2µm filter)

3

Ras, J., Uitz, J., and Claustre, H.: Spatial variability of phytoplankton pigment distribution in the South East Pacific, Biogeosciences Discuss., in preparation, 2007.

(8)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU were put into 20 mL Teflon bottles. 2.5 mL of reagent (140 ml of NaOH 1.5 M, 30 g of

H3BO3, 360 ml of demineralised water) was added and the mineralization processed (autoclave 30 min, 1 bar). After cooling down to ambient temperature, DIP was mea-sured in the same bottles as for mineralization. All reagents were prepared with pro analysis MerckTM Reagent Grade chemicals and with Milli-QTM high purity

deminer-5

alised water. All utensils were washed with 10% hydrochloric acid and rinsed three times with demineralised water.

Particulate organic carbon concentration was determined by the wet-oxidation pro-cedure (Raimbault et al., 1999), following the filtration of 1.2 L of seawater through

0.2µm teflon membranes.

10

Chlorophyll a (Chl a) concentration was determined by the serial filtration of 1 to 1.2 L samples following the same filtration method as for PartP. Immediately after filtration, the filters were put in cryotubes with 5 mL of methanol for pigment extraction (30 min, 4◦_{C) (Herbland et al., 1985). The fluorescence was measured using a Turner designs} 10-AU-005-CE fluorimeter equipped with a chlorophyll a Kit (F4T45.B2 lamp) according

15

to Welschmeyer (1994).

Picophytoplankton (Prochlorococcus, Synechococcus and picophytoeukaryotes) and bacterial abundances were determined according to Grob et al. (2007) using a FACSCalibur (Becton Dickinson) flow cytometer. Picophytoplankton abundances were determined in situ on fresh samples while bacterioplankton samples were fixed with

ei-20

ther paraformaldehyde at 1% or glutaraldehyde at 0.1% final concentration and frozen in liquid nitrogen. Samples were then processed according to Marie et al. (2000a, b). At each sampling depth, corresponding to the Chla and P and C uptake rates measurements, 2 mL samples were filtered through 0.6µm polycarbonate filters. The

filtrate was then analysed using flow cytometry and compared to the total in the

corre-25

sponding sample.

Carbon and phosphate uptakes were determined using the 33P/14C dual labeling method (Duhamel et al., 2006). Duplicate samples (300 ml) were collected into sample-rinsed, polycarbonate bottles (Nalgene) for each sampling depth. An additional

(9)

dupli-BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU cate sample (300 ml) of surface water was incubated with 300µl of HgCl2 (20 g L−1)

to act as a control for non-biological assimilation (Kirkwood 1992). The samples were inoculated with 1080 kBq carrier-free33P (<40 pmol L−1 _{final concentration –} or-thophosphate in dilute hydrochloric acid; Amersham BF 1003; half-life 25.383±0.040 days; Duhamel et al., 2006), and 3.7 MBq14C (bicarbonate aqueous solution;

Amer-5

sham CFA3; half-life 5700±30 years; Duhamel et al., 2006). Samples were incu-bated under simulated conditions for 4 to 5 h. Incubation boxes equipped with light filters (nickel screens) were used to reproduce the light level at the appropriate sam-ple depths (50, 25, 15, 7, 3, 1% of transmitted light). Following incubation, 600µL of

KH2PO4 (10 mmol L−1) was added to each flask in order to stop labelled DIP

assim-10

ilation. Samples were kept in the dark to stop DIC uptake. Fractions of 50 mL were filtered through 25 mm polycarbonate membranes (0.2, 0.6 and 2µm) which had been

placed on GF/F filters soaked with saturated KH2PO4, using a low-pressure suction (<0.2 bars). When all samples were filtered, the pressure was increased to 0.6 bars for

5 s in order to eliminate un-incorporated33P. Filters were placed into scintillation vials

15

(Wheaton low-potassium 6 mL glass-clear vials with screw-cap foil liner) with 150µl

of HCl (0.5 mol L−1) in order to eliminate any un-incorporated 14C. After 12 h, 6 ml of scintillation liquid (Ultimagold MV scintillation liquid, Packard) was added to each vial before the first count. Counting (count per minute – cpm) was carried out on a Packard Tri-Carb®2100TR scintillation counter. In order to separate the activity due to33P from

20

that of14C, we applied the method using the different half-lives of the two isotopes (For more details, see Duhamel et al., 2006). A second count was made after a year, sam-ples having been preserved in the dark at room temperature. C and P uptake rates measurements in each size fraction (0.2–0.6; 0.6–2 and>2 µm) were obtained using

difference calculations.

25

Bacterial production was determined by [3H]-leucine incorporation using the centrifu-gation method (Smith and Azam, 1992) according to Van Wambeke et al. (2007b)1. A factor of 1.5 kg C mol leucine−1 _{was used to convert the incorporation of leucine to} carbon equivalents, assuming no isotopic dilution (Kirchman, 1993).

(10)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU 2.3 Daily rates

The daily C uptake rates have been calculated using the method of Moutin et al. (1999). The model enables a conversion factor to be calculated which permits net hourly DIC uptake rates (nmol L−1_h−1_{) to be transformed into net daily rates (nmol L}−1_d−1_{). The} model takes into account the geographical position (latitude and longitude), the

sam-5

pling date, the time of dawn, incubation start time and the time the incubation ended (GMT). The model of Moutin et al. (1999) that previously took theoretical solar radiation into account has been modified to take into account the surface irradiance measured on board.

Daily P uptake rates may be calculated simply by multiplying the hourly rate by 24.

10

Indeed, in several studies (Perry and Eppley, 1981; Moutin et al., 2002), P uptake was shown to be constant over 24h.

2.4 Specific uptake rate estimates

Specific uptake rates (Vsp) have been calculated by dividing the estimators of produc-tion (heterotrophic bacterial producproduc-tion – HBP, C uptake rates – VDICor P uptake rates

15

– VDIP) by estimators of biomass (Phytoplankton and/or bacteria cytometric counts converted in AB and HBB, respectively, or particulate P – PartP). V_DIPsp corresponds to the VDIP to PartP ratio, V

sp

DIC corresponds to the VDIC to AB ratio, HBP:HBB cor-responds to the HBP to HBB ratio. A conversion factor of 10 fg C cell−1 (Christian and Karl, 1994; Caron et al., 1995) has been used to convert heterotrophic bacterial

20

abundance (counted by flow cytometry) to C equivalent. AB has been calculated us-ing two methods. The first one uses a cell-number-to-biomass conversion factor. We chose the Campbell et al. (1997) estimates for Prochlorococcus, Synechococcus and Picoeukaryotes (Table 1). The second method uses a Chla-to-biomass conversion fac-tor. For stations outside the gyre, we chose 70 g C g Chla−1, the average value found

25

(11)

sta-BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU tions inside the gyre, we used values of Chla-to-biomass conversion factor varying with

PAR: 185, 120, 90, and 20 g C g Chla−1 for 50 and 25%, 15%, 7%, and 3 and 1% of PAR, respectively. These factors were chose according to the results found by Veldhuis and Kraay (2004) at the most oligotrophic station of their transect in the Atlantic tropical gyre.

5

The V_DIPsp have been calculated for four size fractions: 0.2–0.6; 0.6–2; >2 and >0.6 µm. We will develop arguments to show that they correspond to heterotrophic

bacteria (V_DIPsp_<0.6), picophytoplankton (V_DIP0sp _.6−2), nano-micophytoplankton (V_DIPsp_>2) and total phytoplankton (V_DIPsp_>0.6), repectively. TheVsp are expressed as daily rates (d−1) so they are comparable with values found in the literature.

10

3 Results

3.1 Cytometry data

We chose to separate bacteria from phytoplankton by filtrating through 0.6µm filters. In

this way, we determined bacteriaV_DIPsp in the 0.2–0.6µm fraction. To verify the accuracy

of our results, we counted the percentage of bacterial cells that passed through a

0.6-15

µm-filter using flow cytometry. On comparing the total and <0.6 µm sample counts, we

found that on average, 91±10% of the heterotrophic bacteria passed through the

0.6-µm-filter whatever the trophic regime (n=90; all euphotic-layer depth included). This

value was in the same range as values obtained in other studies (∼80%; Obernos-terer et al., 2003). Cytometry data showed that Prochlorococcus (when detectable),

20

Synechococcus and Picoeucaryotes cells had an average size of 0.68±0.08 µm;

0.86±0.1 µm (1.16±0.02 µm for the upwelling stations) and 1.74±0.13 µm all over the transect, respectively (Results from Grob et al., 2007). Nevertheless, cytometry counts showed that 34±24% of the Prochlorococcus cells and 3±5% of the Synechococcus cells were found in the<0.6 µm fraction.

(12)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU 3.2 Evaluation of using particulate phosphate as an estimator of living biomass

Figure 2 shows a typical example of the vertical distribution of PartP and Chla con-centrations compared to the vertical distribution of cell counts by flow cytometry. In the upper 80 m, PartP concentrations were fairly constant, varying between 10.0 and 10.4 nmol L−1 from surface to the depth of 7% PAR. PartP concentrations decreased

5

to 5.3 nmol L−1at the depth of 1% PAR while Chla concentrations increased from 0.05 to 0.19µg L−1 _{from the surface waters down to 1% of PAR, respectively. So, contrary} to Chla, PartP did not show a deep concentration maximum (Fig. 2). Phytoplankton cell counts by flow cytometry showed an increase from 1.1×105to 2.8×105cells mL−1 from surface water to the depth of 3% of PAR and a decrease to 1.3×105 cells mL−1

10

at the depth of 1% of PAR. In the same way, total cytometric counts (bacteria + phy-toplankton) showed an increase from 6.8×105to 8.5×105cells mL−1 from the surface water to the depth of 15% of PAR and a decrease to 6.5×105cells mL−1_{at the depth of} 1% of PAR. Variations in PartP concentration throughout the euphotic zone are closer to that of cell concentration than to Chla concentration.

15

Chla is largely used to estimate phytoplankton biomass (Trembaly and Legendre, 1994; Uitz et al., 2006). Nevertheless, as illustrated by Fig. 2, the C:Chla ratio is subjected to variations depending on light (Taylor et al., 1997). For this reason, only data between 50 and 15% of transmitted light are considered for the comparison of biomass estimates to Chla concentration, to avoid any biases associated with

pho-20

toacclimatation. In order to check whether PartP below 0.6µm is representative of

bacteria and PartP above 0.6µm is representative of phytoplankton biomass, data in

terms of PartP and Chla were compared in the two fractions (Fig. 3). The relation between Chla and PartP in the 0.2–0.6µm fraction was less significant than in the >0.6 µm fraction (r2=0.31, P<0.001; and r2=0.86, P<0.001, respectively; Fig. 3). This

25

indicates that 69% and 14% of the dispersion can not be explained by the regression in the fractions between 0.2–0.6µm and above 0.6 µm, respectively. The contribution

(13)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU but this correlation is comparatively weak and suggests the contributor to PartP are

organisms that do not contain Chla. In other words, it is mainly free-living bacteria that contribute to PartP in the 0.2–0.6µm fraction. In the same way, the essential part of

the PartP in the>0.6 µm fractions was correlated to Chla, supporting the hypothesis

that it corresponds essentially to phytoplankton biomass.

5

The regression slope between POC and PartP concentration was 349 (Fig. 4a) while betweenVDICandVDIP was 57 (Fig. 4b). This indicates that the turnover rates of POC and PartP were different, and supports the hypothesis that P is more rapidly mineral-ized from dead material than C. The correlation between Chla and PartP concentration data was also better (r=0.87, p<0.001, Fig. 4c) than that between Chla and POC

10

(_{r=0.51, p<0.05, Fig. 4d), supporting the idea that PartP was a better indicator of} living biomass than POC. Eutrophic stations have been omitted to avoid regressions being drawn by high values.

3.3 Evaluation of the use ofV_DIPsp as an estimator of bacteria and phytoplankton growth rates

15

We compared euphotic-layer averaged values of Vsp obtained from different methods (Fig. 5). For bacteria, we compared the values ofV_DIPsp_<0.6 and of HBP:HBB (Fig. 5a).

V_DIPsp_<0.6 values were 1.2 to 9.5 times higher than HBP:HBB values in rich areas (MAR to STB6 and STB15 to UPX) while in the centre of the gyre (STB7 to STB14),V_DIPsp_<0.6

values were 1.2 to 2.2 times lower than HBP:HBB values. Between MAR and STB21

20

stations, HBP:HBB values were quite low and constant (0.11±0.04) while V_DIPsp_<0.6 val-ues varied widely from 0.04 to 1.11 d−1 _{depending the trophic regime encountered.} As a consequence, the correlation between HBP:HBB ratio andV_DIPsp_<0.6 values, even excluding the “original” upwelling sites, was not significant (r2=0.08, p>0.05). For phytoplankton, we compared the values of V_DICsp_>0.6 and of V_DIPsp_>0.6. V_DICsp_>0.6 can be

25

obtained using a Chla-to-biomass or cell-number-to-biomass conversion factor. Using cell-number-to-biomass conversion factors according to Campbell et al. (1994, 1997)

(14)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU we foundV_DICsp values on average 2 and 12 times higher than using a Chla-to-biomass

conversion factors. Highest differences were found in meso- and eutrophic areas. With this Chla-to-biomass conversion factor,V_DICsp_>0.6 values were 1 to 4 and 0.6 to 1.8 times higher than V_DIPsp_>0.6 in the gyre and in the meso- and eutrophic areas, respectively (Fig. 5b). The major bias linked with the determination of V_DICsp is the choice of the

5

conversion factor. Indeed, considering extreme values of the C:Chla ratio found in liter-ature for surface layer of the equatorial Pacific ocean (40 and 200 g C Chla−1_{, Chavez} et al., 1996), values ofV_DICsp can vary up to a factor 5. In the same way, we calculated

V_DICsp using different cell-number-to-biomass conversion factor. Using conversion factors provided by Campbell et al. (1997) or by Bertilsson et al. (2003) for Prochlorococcus,

10

Synechococcus and picoeucaryotes (see Table 1), we found thatV_DICsp values were on average 20% higher with Campbell’s value.

3.4 Estimates of bacteria and phytoplanktonV_DIPsp in the Southeast Pacific gyre The different size fractions showed significant vertical and longitudinal variations of

V_DIPsp along the transect (P<0.001; Fig. 6). Highest values were found in rich areas

15

while lower values were found in the upper part of the photic zone in the gyre area. In rich areas, the 0.2–0.6µm fraction, assumed to be composed mostly of free-living

heterotrophic bacteria, showed the highest euphotic zone averaged values of V_DIPsp

(0.6±0.3 to 3±1 d−1_{) while in the gyre area, the 0.6–2}_{µm fraction, assumed to consist} of picophytoplankton cells, showed the highest euphotic zone averaged values ofV_DIPsp

20

(0.10±0.04 to 0.20±0.11 d−1_{). Whatever the station, the}_{>2 µm fraction had the lowest}

V_DIPsp euphotic zone averaged values (0.02±0.07–0.6±0.2 d−1). The variation of V_DIPsp

with depth in the 0.6–2µm fraction was quite different from that of the >2 µm fraction,

particularly in the west part of the gyre area. In the gyre area, while the>2 µm

frac-tion showed quiet constantV_DIPsp values with depth (no significant difference was found

25

(15)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU layer of the euphotic zone, P=0.161), the 0.6–2 µm fraction showed significantly higher

values ofV_DIPsp in the DCML (P<0.001). The smaller size fraction (0.2–0.6 µm) showed

quite low values throughout the euphotic zone in the oligotrophic area (0.11±0.07 d−1₎ and no significant tendency with depth was observed (P<0.001).

4 Discussion

5

Quantifying heterotrophic bacteria and phytoplanktonµ in the ocean is of critical

impor-tance to understand many oceanographic processes sinceµ of individual populations

control the ultimate composition of the assemblage (Banse, 1991). This, in turn, con-trols a large number of ecosystem properties, such as export of organic matter, nutri-ents utilization and production patterns. Knowledge ofµ is critical to our understanding

10

of the biotic responses to environmental forcing. The physiological responses are an integral component in mechanistic models to predict ecosystem trophodynamics. Nev-ertheless, studies of heterotrophic bacteria and phytoplankton assemblages are still scarce, especially in the Southeast Pacific. We measured DIP uptake rates and PartP concentrations in three size fractions: 0.2–0.6, 0.6–2 and >2 µm in order to assess

15

both heterotrophic bacteria and two size fractions of phytoplanktonµ. First, we discuss

the production and biomass estimators; second, we discuss the P-basedµ estimates

obtained in the Southeast Pacific.

4.1 Biomass estimators

The distribution of phytoplankton is commonly described in terms of Chla (Huot et al.,

20

2007). Because the Chla content varies between species and with light and nutrients (Philips et al., 1995; Sciandra et al., 1997; Finkel et al., 2004; P ´erez et al., 2006; Moore et al., 2006), it is not an ideal biomass estimator (Breton et al., 2000; Le Floc’h et al., 2002). POC cannot be used directly because it contains a high proportion of detrital matter (Sobczak et al., 2002; Figs. 4a and d). The AB in terms of C is never directly

(16)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU determined but derived from other variables: Chla, biovolumes or cell numbers which

are then transformed using appropriate conversion factors. This entails a critical step in the estimation of AB: the choice of conversion factor. C:Chla values vary over a wide range (Table 2). For phytoplankton, cell-number-to-C conversion factor can vary signif-icantly (Table 2) even at a species level (Prochlorococcus CCMP 1378: considering C

5

is independent of light = 49±9 fg C cell−1 and considering C content varies with light = 65±67 and 48±10 fg C cell−1, for cultures switched from low light to high light and from high light to low light, respectively – Cailliau et al., 1996; Prochlorococcus PCC 9511 grown in turbidostat under a daily light cycle = 27±6 fg C cell−1 – Clautre et al., 2002; Synechococcus WH8102 grown under continuous white light in batch cultures =

10

279.1±84.2 fg C cell−1 _{– Six et al., 2004). We found significant differences (P<0.001)} inV_DICspestimates according to the choice of phytoplankton cell-number-to-C or C:Chla conversion factors. So, although the employ of conversion factors is still largely used, it is difficult to choose the appropriate one. Because they vary with in situ conditions and species composition, it should be necessary to use different factors for all

sam-15

ples. This would result in a complex data analysis. Studies on heterotrophic bacterial communities have shown that the C cell content changes in relation to natural condi-tions and the physiological state of the bacterial assemblages (Table 2). Gundersen et al. (2002) showed that the outcome of HBB assessments is highly dependant on the choice of cell-specific conversion factors. In the same way, La Ferla and Leonardi

20

(2005) demonstrated that the quantification of HBB based solely on abundance must be considered with caution because of the variability in cell volumes and morphotypes. Therefore, there is great uncertainty surrounding the estimate of C-based phytoplank-ton and heterotrophic bacterialµ, whatever the choice of biomass estimator.

P is an essential element required for life, used by all organisms. It is found in a wide

25

range of molecules with varying cellular roles, going from storage of genetic informa-tion (nucleic acids: DNA, RNA) and energy (ATP, ADP, AMP) to structural composiinforma-tion (phospholipids). If participation of detrital P is low in the PartP stock, then it would represent cellular P content. Our results showed that PartP contained less detrital

(17)

ma-BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU terial than the POC (Fig. 4). This was indicated by significantly higher turnover rates

of PartP compared to those of POC and a significant correlation between PartP and Chla concentrations. Similar observations have shown that P is preferentially released into the water column relative to other elements such as C and N (Menzel and Ryther, 1964; Knauer et al., 1979; Minster and Boulahdid, 1987; Loh and Bauer, 2000; Paytan

5

et al., 2003). The use of PartP as a living biomass indicator is particularly well adapted to the open ocean. Indeed, in such areas, low values of detrital P are commonly found (∼1% in equatorial Pacific Ocean, Faul et al., 2005). Nevertheless, even if the frac-tion of detrital P is negligible in the whole fracfrac-tion, the size distribufrac-tion of detrital P is not known and can affect the measurement of V_DIPsp in each size fraction. It has been

10

shown that the more the size of the organic matter decreases, the more refractory it is (the size-reactivity continuum hypothesis; Amon and Benner, 1996; Mannino and Har-vey, 2000), therefore we can hypothesis that there is also a size-reactivity continuum in detrital matter that engender higher concentration of detrital matter in the smallest frac-tion. In this way,V_DIPsp_<0.6may be underestimated. The proportion of detrital P in PartP is

15

high in coastal areas (Faul et al., 2005). Consequently,V_DIPsp is more likely to be under-estimated in the upwelling area. The other main advantage of using P instead of Chla is that PartP takes both bacteria and phytoplankton into account. So if it is possible to separate bacterial P from phytoplankton P in PartP, then it would be possible to esti-mate bacterial and phytoplanktonV_DIPsp in the same sample. Size fractionation was an

20

adequate method for separating heterotrophic bacteria from phytoplankton since more than 90% of bacterial cells passed through the 0.6µm-filters. However, an increasing

fraction of Prochlorococcus cells passed through this filter when water became ultra-oligotrophic (in the centre of the gyre). Consequently, values of heterotrophic bacteria

V_DIPsp in the gyre may be biased due to the influence of Prochlorococcus cells.

Never-25

theless, it was shown that DIC uptake in the 0.2–0.6µm fraction was negligible (data

not shown) and therefore the phytoplankton production in this fraction was negligible. Thus, production in terms of P in the 0.2–0.6µm fraction can be mainly attributed to

(18)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU bacterialµ. In >0.6 µm size fractions, the nano and microzooplankton can account for

a fraction of the PartP concentration values. Gasol et al. (1997) showed that zooplank-ton C-biomass (protozooplankzooplank-ton + mesozooplankzooplank-ton) could account for 13–21% and 15–65% of the total C-biomass in coastal and open ocean areas, respectively. Conse-quently, this could be a non negligible source of phytoplanktonV_DIPsp underestimation,

5

particularly for the>2 µm size fraction.

4.2 DIP uptake rate measurements

Assuming the DIP to represent biologically available orthophosphate, we measured P uptake rates (VDIP) in three size fractions. Daily P uptake rates were calculated by mul-tiplying the hourly rate by 24. P uptake is generally shown to be constant over 24 h

10

(Perry and Eppley, 1981; Harrison, 1983; Moutin et al., 2002) but diurnal variations in P uptake have been observed in some studies (Eppley et al., 1971; Harrison et al., 1977; Currie and Kalff, 1984). For the majority of stations, time course experiments for 33P uptake were linear over 24 h, however there were some variations in P up-take rates at some stations along the BIOSOPE transect (Duhamel et al., 2006). The

15

methodological problems associated with 24-h incubation experiments can be signifi-cant (Nalewajko and Garside, 1983; Harrison and Harris, 1986), especially in terms of losses. However, short incubation experiments are supposed to reduce the bias linked to such losses (see discussion in Duhamel et al., 2006). Nevertheless, it is important to stress that even if the<0.6 µm fraction is composed of solely heterotrophic bacteria,

20

our data set does not prove that P is turning over at the same rate as the cells. Indeed, Nalewajko and Lean (1978) measured net phosphate uptake and influx rates in batch cultures of three algal cells. They showed that short-term P fluxes always exceeded the net increase in P biomass, indicating that the cells release P compounds back into the medium. To the best of our knowledge, the study of Nalewajko and Lean (1978)

25

has not been repeated, so it would be necessary to repeat this experiment in a variety of field samples in order to verify that this phenomenon is not exclusively observed in cultures. C-basedµ estimations are also submitted to such error type. Indeed, the

(19)

re-BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU lease of assimilation products is common to C measurements. Claustre et al. (2007)2

propose that the release of DOC in the Southeast Pacific should be a major process which could explain the high community (bacteria + phytoplankton) production rates. So, the release of DOC by phytoplankton cells produces bias in the evaluation of C production (Wood et al., 1992) and subsequently for C-basedµ estimations.

5

4.3 Growth rates estimates

In 1981, Perry and Eppley used the 33P uptake rate to PartP ratio to estimate the growth rate of phytoplankton making the statement that DIP assimilation was mediated by phytoplankton (their data indicating low heterotrophic activity). In 1996, Thingstad et al. used the32P uptake rate to PartP ratio to evaluate both heterotrophic bacteria and

10

phytoplankton growth rates using 1µm size fractionations. From these different studies,

where the proportion of detrital matter in the PartP was negligible, it was possible to put forward the hypothesis that DIP was the sole source of P and soV_DIPsp estimates could be used to assess bacteria and/or phytoplanktonµ. Thus the idea of using P-based

estimates of µ is not new. In this study we provide information on the variations in

15

P-basedµ values in a gradient of oligotrophy where the waters where P-repleted (DIP

concentration and turnover time minimum values: 120 nmol L−1and 7 d−1, respectively; Moutin et al., 20074). Most estimates for C-based heterotrophic bacterial growth rates in the open ocean fall into a wide range from zero to 10 d−1_{, whilst phytoplankton} appears to grow at rates of no more than 2 d−1 (Tables 1 and 3). We report a wide

20

range inµ estimates ranging from 0 to 7 d−1 _{for heterotrophic bacteria and from 0 to} 2 d−1 _{for phytoplankton. This range of values reflects the wide range of trophic status} encountered during the BIOSOPE cruise.

Estimates of the production to the biomass ratio, based on the leucine

incorpora-4

Moutin, T., Karl, D., Duhamel, S., Rimmelin, P., Raimbault, P., Van Mooy, B., and Claustre, H.: Phosphate availability and the ultimate control of nitrate input by nitrogen fixation in the Pacific Ocean, Biogeosciences Discuss., in preparation, 2007.

(20)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU tion technique and C conversion of bacterial abundance (HBP:HBB) were significantly

lower than those estimated by V_DIPsp_<0.6 in rich areas (MAR to STB6 and STB15 to UPX). Applying various methods (measurements of the natural abundance of nucleoid-containing cells by combined epifluorescence and phase-contrast microscopy; Detec-tion of the reducDetec-tion of the fluorogenic dye, 5-cyano-2,3-ditolyl tetrazolium chloride;

5

Nucleic acid double staining (SYBR Green + propidium iodide); determination of mem-brane integrity by confocal laser-scanning microscopy), it has been shown that at any given time, a significant fraction of the bacterioplankton community has minimal or no metabolic activity (Zweifel and Hagstr ¨om, 1995; Sherr et al., 1999; Gregori et al., 2001; Pirker et al., 2005). For this reason,µ estimates based on the HBP:HBB ratio could be

10

underestimated. Our V_DIPsp_<0.6 values were significantly higher in the rich areas (MAR to STB6 and STB15 to UPX, P<0.001) than in the gyre area. In the same way,

Mor-gan et al. (2006) found that bacterial growth rates (withµ=HBP/bacterial abundance,

HBP deduced from3H-Thymidine method using conversion factor of 2×1018 cells × [mol TdR]−1) were significantly greater on the shelf (0.8–1.8 d−1) compared to the gyre

15

(0.1–0.3 d−1) in the western Black Sea.

Studies comparing bacterial and phytoplankton µ are scarce (Jones et al., 1996;

Almeida et al., 2002). Measurements of V_DIPsp in the 0.2–0.6, 0.6–2 and >2 µm

frac-tions have enabled us to make such comparisons. In oligotrophic environments, het-erotrophic bacterial µ can be higher or lower than that of phytoplankton. For

exam-20

ple, P ´erez et al. (2006) showed that in the upper water (mixed layer) of the subtrop-ical Atlantic gyres, phytoplankton growth rates were 0.17 d−1 (from daily AP and pi-coplankton abundance transformed to B with the empirical conversion factors obtained by Zubkov et al. (2000), see Table 1). While in the same area, Zubkov et al. (2000) found that heterotrophic bacterial growth rates were 0.12 d−1(using a conversion factor

25

of 11.5 fg C per heterotrophic bacteria). In the upper 40 m of the North Pacific subtrop-ical gyre, Jones et al. (1996) found 0.7 d−1 and ∼1 d−1 for phytoplankton (estimated from the Chla-labelling technique) and heterotrophic bacteria (estimated from the in-corporation of 3H-adenine into DNA), respectively. We showed that

(21)

picophytoplank-BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU tonµ (0.14±0.04 d−1_{) were higher that heterotrophic bacteria}_{µ (0.11±0.07 d}−1_{) in the}

Southeast Pacific gyre and found values in the range than those found in the Atlantic and North pacific gyres by P ´erez et al. (2006) and Zubkov et al. (2000), suggesting the existence of a microbial community that turns over very slowly. These relatively low values ofµ for both phytoplankton and heterotrophic bacteria in the oligotrophic gyre

5

area must be the result of the limitation of both bacterial and primary production by nitrogen (Bonnet et al., 20075; Van Wambeke et al., 2007a6). Slow phytoplanktonµ in

the subtropical Atlantic have been explained in terms of the observed assimilation num-bers and C:Chl-a ratios in a review by Mara ˜non (2005). The light-saturated, chlorophyll normalised photosynthesis rate necessary to support phytoplanktonµ of 1 d−1 _would

10

be well above those reported in the subtropical Atlantic (156±16 and 205±17 mg C m−2_d−1_{, in the North and South Atlantic subtropical gyres, respectively; P ´erez et} al., 2006) which were in the range of those measured in the Southeast Pacific gyre (134±82 mg C m−2d−1; Van Wambeke et al., 2007b1). In coastal areas heterotrophic bacterialµ are often found to be lower than that of phytoplankton (Laws et al., 1984;

15

Revilla et al., 2000). In the Southeast Pacific, we found that organisms in the<0.6 µm

fraction had higher V_DIPsp values than organisms in the >0.6 fraction in the rich areas

(MAR to STB6 and STB15 to UPX), while in the hyperoligotrophic gyre, organisms in the 0.6–2µm fraction yielded the highest V_DIPsp values. Thus it may be deduced that the picophytoplankton was more efficient than nano-microphytoplankton and free living

20

heterotrophic bacteria for growing in such hyperoligotrophic conditions.

There are relatively few studies comparingµ for different size fractions of natural phy-5

Bonnet, S., Guieu, C., Bruyant, F., Prasil, O., Raimbault, P., Gorbunov, M., Zehr, J. P., Grob, C., Masquelier, S., Garczareck, L., Moutin, T., Van Wambeke, F., and Claustre, H.: Nu-trients controlling primary productivity in the South East Pacific, Biogeosciences Discuss., in preparation, 2007.

6

Van Wambeke, F., Bonnet, S., Moutin, T., Raimbault, P., Alarc¸on, G., and Guieu, C.: Factors limiting heterotrophic prokaryotic production in the Southern Pacific Ocean, Biogeosciences Discuss., in preparation, 2007a.

(22)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU toplanktonic communities (P ´erez et al., 2006). In coastal eutrophic ecosystems, large

phytoplankton have been reported to have faster growth rates than the small-sized phytoplankton (Cermeno et al., 2005 – C-specific photosynthetic rates). Neverthe-less, in the Chilean upwelling area, there was no significant difference between V_DIPsp for the two size-fractions of phytoplankton (0.5±0.3 and 0.4±0.2 d−1for picophytoplankton

5

and nano-microphytoplankton respectively,p>0.05). We found that picophytoplankton

(0.6–2µm) was growing 1 to 15 times faster than the nano-microphytoplankton (>2 µm)

between the Marquesas Islands and Chile, with maximal differences in the gyre area. Differences in growth rates have been reported to be related to the specific compo-sition of the planktonic community (Furnas, 1990). So the differences we observed

10

could be related to differences between the taxonomic groups encountered along the BIOSOPE transect. Indeed, flow cytometry data showed high variations in the relative composition of picophytoplankton populations along the BIOSOPE transect (Grob et al., 2007). In the hyperoligotrophic region, the DCM corresponded to Prochlorococcus and picophytoeukaryotes maxima (Grob et al., 2007) as well as the maximum growth

15

rates values of the picophytoplankton size fraction (Fig. 6).

In most of the oligotrophic area, phytoplanktonµ were found to be higher in the upper

mixed layer than within the proximity of the DCML (Malone et al., 1993 – with AP de-duced from the14C labelling method and AB deduced using a C:Chla ratio or14C-Chla experiments ; Quevedo and Anadon, 2001 – dilution method). We found that

picophyto-20

plankton grew significantly faster at the DCML than in the upper part of the photic zone in the hyperoligotrophic gyre area (from STB7 to STB14; P< 0.001). P ´erez et al. (2006)

found the same trends in the subtropical Atlantic gyres withµ in the <2 µm fraction of

0.17±0.01 d−1 in the mixed layer and 0.25±0.02 d−1 in the DCML. Neverthelees, they found that large size fraction (>2 µm) was growing faster in the mixed layer than in the

25

DCML while we found no statistical difference for V_DIPsp_>2(P=0.161). Our results support the hypothesis of P ´erez et al. (2006) that picoplankton might outcompete large cells in high-nutrient, low-light environment of the DCML.

(23)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU to judge which technique is the best for measuringµ, if indeed any one technique is

capable of doing such as each method measures a different aspect of growth. P-based

µ estimates are one of the many ways to assess µ and comparing the results obtained

using the different existing methods can help to understand how fast the cells grow in relation to the area of ocean under study (Christian et al., 1982; Laws et al., 1984;

5

Jespersen et al., 1992).

5 Conclusions

Growth rate is a fundamental property of all organisms and is especially informative about the activity of microbial populations. The relative activity of bacteria and phyto-plankton in oligotrophic oceans has significant implications for the food-web structure,

10

nutrient cycling pathways and for sinking flux of organic matter. Contrary to C-based approaches, the P-based approach allows to assess bacterial and phytoplankton µ

on the same sample, to the extend that size fraction can isolate efficiently both het-erotrophic and phytoplanktonic fractions. We have characterized the vertical and lon-gitudinal variability of P-basedµ in three size fractions of plankton. Picophytoplankton

15

(0.6–2µm) grew faster than the large phytoplankton (>2 µm) all over the Southeast

Pacific transect and particularly in the centre of the gyre. Thus, cells smaller than 2µm

were better adapted for growing in a wide range of trophic conditions than those greater than 2µm. Heterotrophic bacteria (0.2–0.6 µm) showed higher variations in P-based µ with maximum rates in rich areas. Picophytoplankton grew faster than heterotrophic

20

bacteria in the Southeast Pacific gyre with values in the range than those found in the Atlantic and North pacific gyres by P ´erez et al. (2006) and Zubkov et al. (2000), suggesting the existence of a microbial community that turns over very slowly.

Acknowledgements. We express our gratitude to O. Ulloa, G. Alarcon and C. Grob for

pro-viding us cytometry data. We thank T. Bentley for help with improving the English. We also

25

thank the crew of the R/V L’Atalante for outstanding shipboard support operations. D. Tailliez and C. Bournot are warmly thanked for their efficient help in CTD rosette management and

(24)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU data processing. This is a contribution to the BIOSOPE project of the LEFE-CYBER program.

This research was funded by the Centre National de la Recherche Scientifique (CNRS), the Institut des Sciences de l’Univers (INSU), the Centre National d’Etudes Spatiales (CNES), the European Space Agency (ESA), The National Aeronautics and Space Administration (NASA) and the Natural Sciences and Engineering Research Council of Canada (NSERC). This work

5

is funded in part by the French Research and Education council.

References

Almeida, M. A., Cunha, M. A., and Alc ˆantara, F.: Seasonal change in the proportion of bacte-rial and phytoplankton production along a salinity gradient in a shallow estuary, Hydrobiol., 475/476, 251–262, 2002.

10

Amon, R. M. W. and Benner, R.: Bacterial utilization of different size classes of dissolved organic matter, Limnol. Oceanogr., 41, 41–51, 1996.

Arin, L., Moran, X. A. G., and Estrada, M.: Phytoplankton size distribution and growth rates in the Alboran Sea (SW Mediterranean): short term variability related to mesoscale hydrody-namics, J. Plankton Res., 24, 1019–1033, 2002.

15

Aristegui, J., Hernandez-Leon, S., Montero, M. F., and Gomez, M.: The seasonal planktonic cycle in coastal waters of the Canary Islands, Sci. Mar., 65, 51–58, 2001.

Azam, F., Fenchel, T., Field, J. G., Gray, J. S., Meyer-Reil, L. A., and Thingstad, F.: The ecolog-ical role of water-column microbes in the sea, Mar. Ecol. Prog. Ser., 10, 257–263, 1983. Banse, K.: Rates of phytoplankton cell division in the field and in iron enrichment experiments,

20

Limnol. Oceanogr., 36, 1886–1898, 1991.

Banse, K.: Should we continue to measure14C-uptake by phytoplankton for another 50 years?, Limnol. Oceanogr.: Bull., 11, 45–46, 2002.

Becquevort, S., Bouvier, T., Lancelot, C., Cauwet, G., Deliat, G., Egorov, V. N., and Popovichev, V. N.: The seasonal modulation of organic matter utilization by bacteria in the Danube–Black

25

Sea mixing zone, Estuar., Coast. Shelf Sci., 54, 337–354, 2002.

Berman, T.: Multiple isotopic tracer methods in the study of growth, dynamics and metabolic processes in marine ecosystems, in: Primary Productivity in the Sea, edited by: Falkowski, P., p. 213–229, Plenum Publishing, 1980.

Bertilsson, S., Berglund, O., Karl, D. M., and Chisholm, S. W.: Elemental composition of marine

(25)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU Prochlorococcus and Synechococcus: Implications for the ecological stoichiometry of the

sea, Limnol. Oceanogr., 48, 1721–1731, 2003.

Bj ¨orkman, K. M. and Karl, D. M.: Bioavailability of dissoved organic phosphorus in the euphotic zone at Station ALOHA, North Pacific Subtropical Gyre, Limnol. Oceanogr., 48, 1049–1057, 2003.

5

Breton, E., Brunet, C., Sautour, B., and Brylinski, J. M.: Annual variations of phytoplankton biomass in the Eastern English Channel: comparison by pigment signatures and microscopic counts. J. Plankton Res., 22, 1423–1440, 2000.

Brock, T. D.: Microbial growth rates in nature, Bacteriol. Rev., 35, 39–58, 1971.

Buck, K. R., Chavez, F. P., and Campbell, L.: Basin-wide distributions of living carbon

com-10

ponents and the inverted trophic pyramid of the central gyre of the North Atlantic Ocean, summer 1993, Aquat. Microb. Ecol., 10, 283–298, 1996.

Cailliau, C., Claustre, H., Vidussi, F., Marie, D., and Vaulot, D.: Carbon biomass and gross growth rates as estimated from 14C pigment labelling method during photoacclimation in Prochlorococcus CCMP1378), Mar. Ecol. Progr. Ser., 145, 209–221, 1996.

15

Campbell, L., Liu, H., Nolla, H. A., and Vaulot, D.: Annual variability of phytoplankton and bacteria in the subtropical North Pacific Ocean at Station ALOHA during the 1991–1994 ENSO event, Deep-Sea Res. Part I., 44, 167–192, 1997.

Campbell, L., Nolla, H. A., and Vaulot, D.: The importance of Prochlorococcus to communitu structure in the central North Pacific Ocean, Limnol. Oceanogr., 39, 954–961, 1994.

20

Caron, D. A., Dam, H. G., Kremer, P., Lessard, E. J., Madin, L. P., Malone, T. C., Napp, J. M., Peele, E. R., Roman, M. R., and Youngbluth, M. J.: The contribution of microorganisms to particulate carbon and nitrogen in surface waters of the Sargasso Sea near Bermuda, Deep-Sea Res., 42, 943–972, 1995.

Cermeno, P., Maranon, E., Rodriguez, J., and Fernandez, E.: Large-sized phytoplankton

sus-25

tain higher carbon-specific photosynthesis than smaller cells in a coastal eutrophic ecosys-tem, Mar. Ecol. Progr. Ser., 297, 51–60, 2005.

Chavez, F. P., Buck, K. R., Coale, K. H., Martin, J. H., DiTullio, G. R., Welshmeyer, N. A., Jacobson, A. C., and Barber, R. T.: Growth rates, grazing, sinking, and iron limitation of equatorial Pacific phytoplankton, Limnol. Oceanogr., 36, 1816–1833, 1991.

30

Chavez, F. P., Buck, K. R., Service, S. K., Newton, J., and Barber, R. T.: Phytoplankton variabil-ity in the central and eastern tropical Pacific, Deep-Sea Res., 43, 835–870, 1996.

(26)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU of bacterial growth rates in mixed batch cultures, Appl. Environ. Microbiol., 43, 1160–1165,

1982.

Christian, J. R. and Karl, D. M.: Microbial community structure at the U.S. -Joint Global Ocean flux Study Station ALOHA: Inverse methods for estimating biochemical indicator ratios, J. Geophys. Res., 99(C7), 14 269–14 276, 1994.

5

Clark, L. L., Ingall, E. D., and Benner, R.: Marine organic phosphorus cycling: Novel insights from nuclear magnetic resonance, Amer. J. Sci., 2999, 724–737, 1999.

Claustre, H., Kerherve, P., Marty, J. C., Prieur, L., Videau, C., and Hecq, J. H.: Phytoplankton Dynamics Associated with a Geostrophic Front – Ecological and Biogeochemical Implica-tions, J. Mar. Res., 52, 711–742, 1994.

10

Cloern, J. E., Alpine, A. E., Cole, B. E., and Heller T.: Seasonal changes in the spatial distri-bution of phytoplankton in small, temperate-zone lakes, J. Plankton Res., 14, 1017–1024, 1992.

Cochlan, W.: The heterotrophic bacterial response during a mesoscale iron enrichment exper-iment (IronEx II) in the eastern equatorial Pacific Ocean, Limnol. Oceanogr., 46, 428–435,

15

2001.

Cotner, J. B. and Biddanda, B. A.: Small players, large role: microbial influence on biogeo-chemical processes in pelagic aquatic ecosystems, Ecosystems, 5, 105–121, 2002.

Currie, D. J. and Kalff, J.: The relative importance of bacterioplankton and phytoplankton in phosphorus uptake in freshwater, Limnol. Oceanogr., 29, 311–321, 1984.

20

Dickson, M.-L. and Wheeler, P. A.: Nitrate uptake rates in a coastal upwelling regime: A com-parison of PN-specific, absolute, and Chl a-specific rates, Limnol. Oceanogr., 40, 533–543, 1995.

Ducklow, H. W.: Production and fate of bacteria in the oceans, Bioscience, 33, 494–501, 1983. Duhamel, S., Zeman, F., and Moutin, T.: A dual-labeling method for the simultaneous

measure-25

ment of dissolved inorganic carbon and phosphate uptake by marine planktonic species, Limnol. Oceanogr. Meth., 4, 416–425, 2006.

Eppley, R. W.: Temperature and phytoplankton growth in the sea, Fish. Bull., 70, 1063–1085, 1972.

Eppley, R. W., Carlucci, A. F., Holm-Hansen, O., Kiefer, D., McCarthy, J. J., Venrick, E., and

30

Williams, P. M.: Phytoplankton Growth and Composition in Shipboard Cultures Supplied with Nitrate, Ammonium, or Urea as the Nitrogen Source, Limnol. Oceanogr., 16(5), 741–751, 1971.

(27)

BGD

4, 2027–2068, 2007

Full Screen / Esc

EGU Eppley, R. W.: Relations between nutrient assimilation and growth rate in phytoplankton with a

brief review of estimates of growth rate in the ocean, in: Physiological Bases of Phytoplank-ton Ecology, edited by: Platt, T., Can. Bull. Fish. Aquat. Sci., 210, 251–263, 1981.

Fabricius, K., Yahel, G., and Genin, A.: In situ depletion of phytoplankton by an azooxanthellate soft coral, Limnol. Oceanogr., 43, 354–356, 1998.

5

Faul, K. L., Paytan, A., and Delaney, M. L.: Phosphorus distribution in sinking oceanic particu-late matter, Mar. Chem., 97, 307–333, 2005.

Finkel, Z. V., Irwin, A. J., and Schofield, O.: Resource limitation alters the 3/4 size scaling of

metabolic rates in phytoplankton, Mar. Ecol. Progr. Ser., 273, 269–279, 2004.

Fuhrman, J. A. and Azam, F.: Thymidine incorporation as a measure of heterotrophic

bacteri-10

oplankton production in marine surface waters: Evaluation and field results, Mar. Biol., 66, 109–120, 1982.

Fuhrman, J. A. and Azam, F.: Bacterioplankton secondary production estimates for coastal waters of British Colombia, Antarctica and California, Appl. Environ. Microbiol., 39, 1085– 1095, 1980.

15

Fukuda, R., Ogawa, H., Nagata, T., and Koike, I.: Direct determination of carbon and nitrogen contents of natural bacterial assemblages in marine environments, Appl. Environ. Microbiol., 64, 3352–3358, 1998.

Furnas, M. J.: In situ growth rates of marine phytoplankton: approaches to measurement, community and species growth rates, J. Plankton Res., 12, 1117–1151, 1990.

20

Gasol, J. M., Del Giogio, P., and Duarte, C. M.: Biomass distribution in marine planktonic communities, Limnol. Oceanogr., 42, 1353–1363, 1997.

Gasol, J. M., Del Giogio, P., Massana, R., and Duarte, C. M.: Active versus inactive bacteria: size-dependence in a coastal marine plankton community, Mar. Ecol. Progr. Ser., 128, 91– 97, 1995.

25

Goericke, R.: Response of phytoplankton community structure and taxon-specific growth rates to seasonally varying physical forcing in the Sargasso Sea off Bermuda, Limnol. Oceanogr., 43, 921–935, 1998.

Gr ´egori, G., Citterio, S., Ghiani, A., Labra, M., Sgorbati, S., Brown, S., and Denis, M.: Resolu-tion of viable and membrane-compromised bacteria in freshwater and marine waters based

30

on analytical flow cytometry and nucleic acid double staining, Appl. Environ. Microbiol., 67, 4662–4670, 2001.